North Korea and the US Nuclear Umbrella: The Credibility Gap

By: Kevin Kallmyer

The CSIS recently hosted the program “North Korea and the U.S. Nuclear Umbrella: Extended Deterrence in East Asia.” The program discussed the utility of U.S. extended deterrence in East Asia and its effectiveness to deter and contain an increasingly aggressive North Korea.

A panel of experts on U.S. nuclear policy and East Asia gathered at the CSIS to lead the discussion, composed of Dr. Patrick Morgan (University of California, Irvine Peace & Conflict Program Tierney Chair), Robert Carlin (National Committee on North Korea Co-Chair and Stanford University Center for International Security and Cooperation Visiting Fellow), Dr. Victor Cha (CSIS Korea Chair and Georgetown University Asian Studies Director), and Jofi Joseph (U.S. Department of State Senior Advisor to Under Secretary of State for Arms Control and International Security), and was moderated by Sharon Squassoni (CSIS Proliferation Prevention Program Director and Senior Fellow).

This is the first of three posts that will discuss this event and U.S. extended nuclear deterrence in East Asia more broadly.  

U.S. extended deterrence strategy in East Asia is unable to fulfill the majority of US policy goals in the region, argued Dr. Patrick Morgan. While Morgan is generally a strong proponent of nuclear deterrence – which he believes is a bedrock factor in promoting international stability – that is not the case for East Asia. Morgan argued that extended nuclear deterrence is the wrong tool to secure regional stability and deter North Korean aggression.

Extended deterrence is at the heart of a slew of US foreign policy goals. Morgan went so far as to compare the US alliance system to a community, and extended deterrence as part of the fabric that weaves this community together. Morgan presented five core goals of extended deterrence in US grand strategy:

1. Protect allies
2. Reassure allies
3. Provide regional security management
4. Constrain allies—limit allied nuclear and conventional weapons proliferation
5. Provide the foundation for US power projection

The expected result of extended deterrence, through the fulfillment of these broad goals, is a security environment that is both stable and durable. A large reason for this, Morgan argued, is that the security foundation created by extended deterrence allows states to change their nature and role in the region without generating incentives for conflict. This point seems validated by 20th century East Asia power politics. Over the past 50 years there have been dramatic power transitions in East Asia. The role and relative power of both China and Japan have changed substantially, but major state conflict has remained absent from the region, despite a plethora of potential catalysts. Morgan would attribute this stability to extended deterrence – actors know that even though a competing state may become powerful, those competitors are not existential threats because the totality of U.S. power is committed to prevent aggression and instability in the region (to read further on his argument it is worth checking out the book Complex Deterrence, edited by Morgan himself) 

However, if we apply these goals to North Korea, it becomes clear that U.S. extended deterrence has achieved very little. While a second Korean war has not materialized (which merely begs the question of the probability a second Korean war would have materialized in the absence of the U.S. umbrella), the United States has not resolved the nuclear issue, prevented missile and nuclear tests, export of this technology to other states, or deterred low-level acts of aggression, such as the Cheonan incident.

Morgan pointed out a host of reasons for this apparent failure. Major factors that contribute to this are: (1) lack of targeted goals—instead the United States tries to do too much with extended deterrence, (2) Chinese policy that undermines U.S. pressure on North Korea, and (3) the absence of viable coalition parties if the United States engaged in combat operations with North Korea. However, Morgan included one rationale that deserves further elaboration: U.S. reliance on its nuclear capabilities to fulfill its extended deterrence commitments.

The United States is engaged in two wars led by a president determined to reduce the salience of nuclear weapons. The result: nuclear threats are no longer credible. The argument, however, goes further than that (as I’m sure Morgan would have illustrated if time permitted). While U.S. extended deterrence is manifested through the whole range of U.S. power and capabilities, extended deterrence is increasingly conflated with extended nuclear deterrence, merely one part, albeit a large one, of U.S. deterrence strategy. This conflation and the reliance on the nuclear arsenal for East Asian extended deterrence is counter-productive for U.S. North Korea policy. 

First, it is important to expand on Morgan’s core point. It is extremely unlikely that the United States would ever use nuclear weapons against North Korea. Use of nuclear weapons would have to be part of a larger military effort against North Korea (it’s simply hard to imagine a scenario where the United States uses nuclear weapons against North Korea without initiating a conflict that gets U.S. forces in South Korea entangled), and this is extremely unlikely. Given the lessons learned from Iraq and Afghanistan, the potential for military overstretch, and the massive political costs from a public weary of collateral damage and war in general, the prospects for war with North Korea are dismissive. 

Further, this particular administration is less likely than any to transform cloaked nuclear threats into reality. Obama has embraced the Prague vision, and as Morgan pointed out, an unfortunate consequence of the non-proliferation agenda is that the credibility of nuclear threats decreases. I doubt Obama is eager to be the first president since Truman to have a section of his biography dedicated to the decision to drop the bomb. 

Even more important, there is no tactical reason for the United States to use its nuclear arsenal against North Korea. U.S. conventional superiority is so overwhelming that, as Morgan put it, it is unlikely we would use anything else. North Korea’s military budget is a mere fraction of the United States’; the United States military budget rivals the combined military budgets of every other country on earth. It simply does not match reality that for the first time since WWII, in a war against North Korea, the United States would believe it needed nuclear weapons to effectively prosecute the war. Even further, the United States is in the final stages of constructing a new conventional weapon, the Massive Ordnance Penetrator (MOP), which could likely reach any protected underground bunker, eliminating the rationale that the US would require nuclear weapons for effective “earth-penetrating” weapons. 

However, there is an additional reason why extended nuclear deterrence is no longer credible. Peter Hayes, the director of the Nautilus Institute and an International Relations professor at Melbourne Institute of Technology, claims in his paper “Extended Nuclear Deterrence, Global Abolition and Korea” that North Korean acquisition of nuclear weapons resulted in the loss of extended nuclear deterrence credibility. Hayes argues that throughout the Cold War, and since, U.S. security policy in East Asia has been dramatically intertwined with U.S. nuclear primacy in the region. Nuclear primacy, however, was not understood as merely the superiority of the U.S. arsenal but as “the guarantee that nuclear proliferation by adversaries such as North Korea…would be halted by the United States.” Unfortunately, the connection between extended nuclear deterrence with preventing North Korean nuclear proliferation was short-sighted, as Hayes describes:

Henceforth, the credibility of US END with allies in this region was tied up directly with the United States’ ability to stop (not merely contain by deterrence) the DPRK’s acquisition of nuclear weapons and use of nuclear threat to compel the United States and others to negotiate with it—what I term the DPRK’s “stalker strategy.” As a result of nearly two decades of slow motion nuclear wrestling with the DPRK culminating in its successful nuclear test in 2009, the credibility of US END has fallen to an all-time low. 

If nuclear threats are no longer credible, and as a result, extended nuclear deterrence is no longer effective, then the question becomes: how should the United States respond to accomplish its broader extended deterrence goals? One important caveat, something that Morgan alluded to, is that the first response should be to scale back the broad and over-arching goals for extended deterrence. Simply put, the United States and its allies have high expectations for what deterrence can achieve with regards to North Korea, and realistic assessment of what can be accomplished never hurts.

However, there is potentially a route that could improve U.S. extended deterrent credibility: honesty. If the Obama administration was honest in the reality that the United States is not going to use nuclear weapons against North Korea and proceeded to limit its extended deterrence to non-nuclear options then it could boost the credibility of its non-nuclear threats. Essentially, if you continually threaten someone with an unloaded gun they are not going to believe your threats, even if you have other credible options to annihilate them. If instead, the United States starts with threats based on its usable, battle tested and existential conventional arsenal, while constraining the potential for nuclear use, then I’d bet the prospects to compel adversaries improves. Michael Gerson of the Center for Naval Analysis, while speaking at the Carnegie Endowment, argued that limiting threats to non-nuclear options can actually benefit deterrence:

My argument comes from a position that a fundamental tenet of deterrence is that limiting your options can in fact enhance deterrence and make you safer. This notion of deliberately tying one’s hands or limiting one’s options is of course attributed to the work of Tom Schelling who argued that limiting one’s own options could be a commitment tactic to enhance the credibility of one’s threats.

The potential to increase the credibility of U.S. extended deterrence by turning away from reliance on extended nuclear deterrence, however, ties into the point Morgan made at the beginning of his presentation. The U.S. alliance system is like a community, and rightly or wrongly, extended nuclear deterrence has been a significant part of that community for a long time. Policy prescriptions for U.S. nuclear policy cannot avoid this fact. Extended nuclear deterrence has symbolic importance to these alliances, and the possibility of limiting the United States to non-nuclear threats must deal with these realities.

Luckily, this is a question that the panel discussed, and will be the topic of the next post in this series.

For those that are interested in reading more of Morgan’s thoughts on the nuclear umbrella in East Asia, I recommend checking out his paper “Considerations Bearing on a Possible Retraction of the American Nuclear Umbrella Over the ROK.”